The carbohydrate-sesquiterpene interface. Directed synthetic routes to both (+)- and (-)-fomannosin from D-glucose

Article abstract of DOI:10.1021/jo8004233

(Chemical Equation Presented) An enantiodivergent strategy for the total chemical synthesis of both naturally occurring (+)-fomannosin (1) and its (-)-antipode (ent-1) from α-D-glucose has been developed and successfully implemented. The key steps in the

Full text of DOI:10.1021/jo8004233

The Carbohydrate-Sesquiterpene Interface. Directed Synthetic
Routes to Both (+)- and (-)-Fomannosin from D-Glucose
Leo A. Paquette,* Xiaowen Peng, Jiong Yang,† and Ho-Jung Kang‡
EVans Chemical Laboratories, The Ohio State UniVersity, Columbus, Ohio 43210-1185
paquette@chemistry.ohio-state.edu
ReceiVed March 3, 2008
An enantiodivergent strategy for the total chemical synthesis of both naturally occurring (+)-fomannosin
(1) and its (-)-antipode (ent-1) from R-D-glucose has been developed and successfully implemented.
The key steps in the overall pathway include the following: (i) application of the zirconocene-mediated
ring contraction of vinyl furanosides for the construction of highly substituted cyclobutanols; (ii) the use
of ring-closing metathesis to form the pendant five-membered ring; (iii) making recourse to a monothio
malonic ester to allow for chemoselective reduction to sensitive lactone intermediate 45; (iv) hydroxyl-
directed dihydroxylation with OsO4 to generate 48; and (v) sequential elimination via a cyclic sulfite and
a cyclobutyl triflate. The bridge between the enantiomeric series consisted of a six-step linkup involving
the structural modification of 22 so as to generate ent-30b. Optical activity was fully preserved throughout.
Introduction
In 1967, Bassett and co-workers succeeded in growing from
still cultures of the wood-destroying Basidiomycetes fungus
Fomes annonsus (Fr.) Karst a sesquiterpene metabolite named
fomannosin.1 F. annonsus causes the death of host cells and
does so prior to hyphal invasion. The pathogen involved was
ultimately shown to be 1, the toxicity of which toward Pinus
tadea seedlings and select bacteria aroused considerable eco-
nomic concern, principally in the southeastern United States.
Since pure 1 had been characterized as an unstable semisolid,
definitive structural characterization was initially achieved by
X-ray diffraction analysis involving the p-bromobenzoylurethane
of the 5,6-dihydro derivative.2 Ten years later, the absolute
configuration of fomannosin was established by a second
crystallographic study involving the (-)-camphanate ester of
dihydro-1.3
13C-labeling experiments have implicated a biosynthetic
pathway in which mevalonate is first converted to humulene
via intermolecular cyclization of trans,trans-farnesyl pyrophos-
phate. Subsequent ring closure to the protoilludyl cation
presumablysetsthestageforoxidativecleavageofacarbon-carbon
bond to afford fomannosin (Scheme 1).4 Although relatively
small in size and endowed with only two stereogenic centers, 1
represents a significant challenge to synthetic chemists because
of its inherent instability under both acidic and basic conditions,
its methylenecyclobutene substructure, and its other unprec-
edented architectural features. The Matsumoto group was the
first to engage in a quest of 1, but their photochemical [2 + 2]
† Current address: Department of Chemistry, Texas A&M University, College
Station, TX 77842.
‡ On sabbatical leave from the Department of Chemistry, Kyung Hee
University, 1 Hoe Ki-Dong, Dong Dae Moon-Gu, Seoul 130-701, Korea.
(1) Bassett, C.; Underwood, R. T.; Kepler, J. A.; Hamilton, P. B. Phytopa-
thology 1967, 57, 1046.
(2) Kepler, J. A.; Wall, M. S.; Mason, J. E.; Bassett, C.; McPhail, A. T.;
Sim, G. A. J. Am. Chem. Soc. 1967, 89, 1260.
(3) Cane, D. E.; Nachbar, R. B.; Clardy, J.; Finer, J. Tetrahedron Lett. 1977,
4277.
4548 J. Org. Chem. 2008, 73, 4548–4558
10.1021/jo8004233 CCC: $40.75 2008 American Chemical Society
Published on Web 05/20/2008

The Carbohydrate-Sesquiterpene Interface
SCHEME 1
SCHEME 3
SCHEME 2
appropriate dicarbonyl precursor, or ring-closing metathesis of
a suitably substituted diene generated from 6.
The acquisition of this cyclobutane, which features three
contiguous stereogenic centers along the periphery of its four-
membered ring, was to exploit the capability of zirconocene9
to deoxygenate 7stereoselectively.10 This tactic would allow for
the generation of 7 from D-glucose and provide desirable linkage
of the carbohydrate realm to the sesquiterpene domain.
route was arrested at the stage of 2.5 This development was
followed closely by the Kasugi/Uda approach to (()-dihydro-
fomannosin acetate (3).6 In 1981, Semmelhack et al. reported
an innovative synthesis of (()-1.7 This hallmark development
constitutes the only successful route to the target phytopathogen
disclosed to the present time. In this full paper, we describe the
evolution of a chemical strategy based upon the zirconocene-
mediated ring contraction of a vinylated furanoside derived from
R-D-glucose that has culminated in the companion total syn-
theses of (+)- and (-)-fomannosin.8
Results and Discussion
Vinylated Furanoside Preparation and the Ring-Contrac-
tion Reaction. The synthesis commenced with the known
tosylate 8, which is readily available in enantiomerically pure
form from D-glucose.12 Selective hydrolysis of one of the two
acetonide units in 8 with acidified methanol13 generated a vicinal
diol whose glycolic C-C bond was oxidatively cleaved with
sodium periodate (Scheme 3). The resulting aldehyde was
oxidized further to the methyl ester. When attempts to effect
this transformation directly with bromine in methanol14 proved
unsuccessful, recourse was made to sequential treatment with
sodium chlorite15 and diazomethane. For the purpose of deoxy-
genation at C3, 9 was subjected to E2 elimination in the presence
of DBU, thereby making possible stereocontrolled catalytic
hydrogenation to deliver 10. Saturation exclusively from the ꢀ
face takes place because steric blockade by the acetonide ring
prohibits approach from the alternative direction. The efficiency
of this five-step sequence is noteworthy in that chromatographic
purification needs to be performed only once and the overall
yield is 90%.
Retrosynthetic Analysis
The delicate nature of fomannosin directed us to consider
installation of one or both of its constituent double bonds at a
late stage. With this in mind, the retrosynthetic plan given in
Scheme 2 was devised. Thus, the unsaturated center linking C5
to C6 was to be masked in the form of a protected hydroxyl
group, with further disconnection across the C2-C4 double
bond in 4 to lead back to intermediate 5. In the constructive
direction, ring-closing metathesis and intramolecular Knoev-
enagel condensation were regarded to constitute two processes
having the potential for crafting the R,ꢀ-unsaturated lactone ring
of the target. For the sake of maximum convergency, construc-
tion of the cyclopentanone subunit could take possible advantage
of several options. These included, but were not limited to, [2
+ 2] cycloaddition of a suitably reactive ketene to 6 followed
by ring expansion, McMurry or related pinacol coupling of an
(9) Negishi, E.; Cederbaum, F. E.; Takahashi, T. Tetrahedron Lett. 1986,
27, 2829.
(4) (a) Cane, D.; Nachbar, R. B. Tetrahedron Lett. 1976, 2097. (b) Cane,
D.; Nachbar, R. B. J. Am. Chem. Soc. 1978, 100, 3208.
(5) Miyano, K.; Ohfune, Y.; Azume, S.; Matsumoto, T. Tetrahedron Lett.
1974, 1545.
(10) (a) Ito, H.; Motoki, Y.; Taguchi, T.; Hanzawa, Y. J. Am. Chem. Soc.
1993, 115, 8835. (b) Hanzawa, Y.; Ito, H.; Taguchi, T. Synlett 1995, 299.
(11) (a) Paquette, L. A.; Cuniere, N. Org. Lett. 2002, 4, 1927. (b) Paquette,
L. A.; Kim, I. H.; Cuniere, N. Org. Lett. 2003, 5, 221.
(6) (a) Kosugi, H.; Uda, H. Bull. Chem. Soc. Jpn. 1980, 53, 160. (b) Kosugi,
H.; Uda, H. Chem. Lett. 1977, 1491.
(12) (a) Schmidt, O. T. Methods Carbohydr. Chem. 1963, 2, 320. (b) Hwang,
C. K.; Li, W. S.; Nicolaou, K. C. Tetrahedron Lett. 1984, 25, 2295.
(13) Just, G.; Luthe, C. Can. J. Chem. 1980, 58, 2286.
(7) (a) Semmelhack, M. F.; Tomoda, S.; Nagaoka, H.; Boelter, S. D.; Hurst,
K. M. J. Am. Chem. Soc. 1982, 104, 747. (b) Semmelhack, M. F.; Tomoka, S.
J. Am. Chem. Soc. 1981, 103, 2427. (c) Semmelhack, M. F. Strategies Tactics
Org. Synth. 1984, 201–222.
(14) Williams, D. R.; Klingler, F. D.; Allen, E. E.; Lichtenthaler, F. W.
Tetrahedron Lett. 1988, 29, 5087.
(8) (a) Paquette, L. A.; Peng, X.; Yang, J. Angew. Chem., Int. Ed. 2007, 46,
7817. (b) Paquette, L. A.; Kang, H.-J. Tetrahedron 2004, 60, 1353.
(15) Hase,T.; Wa¨ha¨la¨, K. Encyclopedia of Reagents for Organic Synthesis;
Paquette, L. A., Ed.; Wiley: Chichester, 1995; Vol. 7, pp 4533-4534.
J. Org. Chem. Vol. 73, No. 12, 2008 4549

Paquette et al.
SCHEME 4
SCHEME 5
SCHEME 6
We next directed our efforts toward more advanced func-
tionalization of furanoside 10. During initial exploratory studies,
it was noted that its hydrolysis with hydrochloric acid in
methanol and subsequent exposure to p-methoxybenzyl trichlo-
roacetimidate16 afforded a 3.5:1 mixture of anomers 11 and 12
in 88% yield. The minor component could be readily epimerized
by heating to 55 °C in a solvent system constituted of methanol
and trimethyl orthoformate containing a catalytic quantity of
HCl. As expected on steric grounds, alkylation of the lithium
enolate of 11 with monomeric formaldehyde17 gave rise to
diastereoisomers 13 and 14 in a ratio of 1.7:1 after hydroxyl
protection as the tert-butyldiphenylsilyl ethers (Scheme 4).18
Although this pair of intermediates proved chromatographically
separable and amenable to efficient conversion to 15 and 16,8b
the generation of anomers at each of two steps was not
considered time-efficient.
Consequently, the hydroxymethylation was performed on 10
at an earlier stage (Scheme 5). By taking advantage of the steric
screening offered by the acetonide substructure, we were able
to achieve the exclusive production of 17 although in more
modest yield. Sequential Dibal-H reduction and Wittig olefi-
nation of 17 to afford 18 proceeded in 81% yield without any
concern for epimerization. Hydrochloric acid-promoted opening
of the acetonide ring in 18 followed by base-promoted coupling
to p-methoxybenzyl bromide delivered the anomeric methyl
glycosides 15 and 20.
The time had now arrived to examine the zirconocene-
mediated ring contractions. While several successful examples
featuring vinylated furanosides as substrates have appeared,10,11
15 and 20 represent members of a modified structural class
where conversion to a cyclobutane would be accompanied with
the generation of a quaternary center. This structural feature
proved not to be an impediment to this reaction, as controlled
exposure of 15 to “Cp2Zr” gave rise in 85% yield to a
diastereomeric mixture of 21 and 22 (2.4:1) (Scheme 6). The
relative stereochemistries of the two products were unequivo-
cally assigned on the basis of NOESY experiments performed
on the corresponding TBS ethers, which were similarly
prepared.8b Rationalization of the product distribution rests upon
consideration of the ring-closure transition states. Initial coor-
dination of the zirconium atom to the vinylic double bond and
furanoside oxygen guarantees E geometry for the developing
allylzirconium intermediate. Advancement along the pathway
to transition state A eventuates in the formation of 21, while
passage through transition state B leads to 22. Presumably, the
adoption of B is less favored as a consequence of existing steric
interaction between the methylene group of the developing
cyclobutane and the allylic methylene group R to the zirconium,
as well as dipole-dipole interactions between the oxygen atoms
resident in the p-methoxybenzyl and tert-butyldiphenylsilyl
ethers. Thus, 22 arises as the less dominant product. When C4-
(16) (a) Overman, L. E. Acc. Chem. Res. 1980, 13, 218. (b) Patil, V. J.
Tetrahedron Lett. 1996, 37, 1481.
(17) Schlosser, M.; Jenny, T.; Guggisberg, Y. Synlett 1990, 704.
(18) The NOE data on which the stereochemical assignments rest have been
provided earlier.8b
4550 J. Org. Chem. Vol. 73, No. 12, 2008

The Carbohydrate-Sesquiterpene Interface
SCHEME 7
soon uncovered that keto aldehyde 28 was a rather sensitive
substance. Following a series of experiments designed to
maximize its availability, recourse was eventually made to
dihydroxylation with 1 equiv of OsO4 and subsequent oxidative
cleavage of the glycolic C-C bond with sodium periodate. We
were disappointed to find, however, that no trace of the
intramolecular cyclization product 29 was found when 28 was
subjected to low-valent titanium- or samarium(II) iodide-based
reductive conditions.
SCHEME 8
To avoid the above problem while maintaining a closely
related tactical approach, ring-closing olefin metathesis23 was
selected as a potential means for construction of the five-
membered ring. The original Grubbs catalyst C24 is known to
be sensitive to the structural environment and is not particularly
effective for the formation of CdC double bonds in crowded
surroundings. Although Schrock’s catalyst D25 is more active,
it is less compatible with select functional groups of the type
presently at hand. In contrast, the second-generation Grubbs
catalyst E has been reported to be both tolerant of a wide range
of functional groups and highly effective in the elaboration of
tetrasubstituted olefinic bonds.26 Our expectation was that E
would prove effective in the fomannosin setting, where instal-
lation of a trisubstituted double bond adjacent to a quaternary
center must be accomplished (Scheme 9).
Before proceeding, the hydroxyl group in 21 was protected
as its methoxyethoxymethyl (MEM) ether.27 The most effective
way to accomplish this task involved coupling to MEM chloride
in the presence of tetrabutylammonium chloride. The conversion
of 21 to 31a via 30a profited from our earlier experience with
this series of reactions. However, the carbonyl methylenation
associated with advancement to 32a proved to be problematic.
Conventional tools for effecting this transformation, including
the Wittig,28 Tebbe,29 and Nysted reagents,30 proved to be
uniformly ineffective, likely due to steric congestion imposed
by the neighboring quaternary carbon. A modicum of success
was achieved in the form of 1,2-addition with methyllithium
followed by dehydration with thionyl chloride. However, this
approach gave 32a in only 24% yield and was accompanied by
epimer 16 was treated in parallel fashion, very similar yields
and product distributions were realized. This outcome is
consistent with the proposed mechanistic pathway in which A
and B are again involved in formation of the cyclobutanols.
Assembly of the Cyclopentanone Ring. The critical role to
be played by the vinyl substituent in 21 was to serve as the
cornerstone of the cyclopentanone subunit (constituted of
C9-C15) in fomannosin (1). Several possible convergent
solutions to this challenge suggested themselves. Of these, three
approaches were accorded attention, the first of which consisted
of dichloroketene in a [2 + 2] cycloaddition strategy followed
by one-carbon ring expansion. However, this ploy was aban-
doned when extensive efforts19 to induce the envisioned
generation of 24 from differently protected 23 were universally
met with failure (Scheme 7).
We therefore turned next to adaptation of the McMurry
reaction20 or related pinacol couplings21 for elaboration of the
cyclopentanone subunit. To this end, cyclobutanol 21 was
initially protected as a methoxymethyl ether under acidic
conditions22 (Scheme 8).
Homologation of the vinyl side chain was achieved via
ozonolysis and subsequent union with the lithium reagent
derived from 5-iodo-4,4-dimethylpent-1-ene. Oxidation of the
resulting carbinol with pyridinium dichromate in the presence
of 4 Å molecular sieves delivered ketone 27 efficiently. It was
(23) (a) Reviews: Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34,
18. (b) Fu¨rstner, A. Angew. Chem., Int. Ed. 2000, 39, 3012. (c) Roy, R.; Das,
S. K. Chem. Commun. 2000, 519. (d) Philips, A. J.; Arbell, A. D. Aldrichchim.
Acta 1999, 32, 75. (e) Armstrong, S. K. J. Chem. Soc., Perkin Trans. 1 1998,
371. (f) Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413. (g) Ivin, K. J. J.
Mol. Catal. A 1998, 133, 1. (h) Alkene Metathesis in Organic Synthesis; Fu¨rstner,
A., Ed.; Springer: Berlin, 1998. (i) Randall, M. L.; Snapper, M. L. J. Mol. Catal.
A 1998, 113, 29. (j) Schuster, M.; Blechert, S. Angew. Chem., Int. Ed. Engl.
1997, 36, 2036.
(19) (a) Danheiser, R. L.; Savariar, S.; Cha, D. D. Org. Synth. 1990, 68, 32.
(b) Johnston, B. D.; Czyzervska, E.; Oehlschlager, A. C. J. Org. Chem. 1987,
52, 3693. (c) Mehta, G.; Rao, H. S. P. Synth. Commun. 1985, 991. (d) Krepski,
L. R.; Hessner, A. J. Org. Chem. 1978, 43, 2879. (e) Greene, A. E.; Luche,
M. J.; Depres, J.-P. J. Am. Chem. Soc. 1983, 105, 2435.
(24) (a) Schwab, P.; France, M. B.; Ziller, J. W.; Grubbs, R. H. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 2039. (b) Schwab, P.; Grubbs, R. H.; Ziller, J. W. Angew.
Chem., Int. Ed. Engl. 1996, 118, 100.
(25) (a) Schrock, R. L. R.; Murdzek, J. S.; Bazan, G. C.; Robbins, J.; Dimare,
M.; O’Regan, M. B. J. Am. Chem. Soc. 1990, 112, 3875. (b) Bazan, G. C.;
Khosravi, E.; Schrock, R. R.; Feast, W. J.; Gibson, V. C.; O’Regan, M. B.;
Thomas, J. K.; Davis, W. M. J. Am. Chem. Soc. 1990, 112, 8378. (c) Bazan,
G. C.; Oskam, J. H.; Cho, H. N.; Park, L. Y.; Schrock, R. R. J. Am. Chem. Soc.
1991, 113, 6899.
(20) (a) McMurry, J. E.; Lectka, T.; Rico, J. G. J. Org. Chem. 1989, 54,
3748. (b) McMurry, J. R. Chem. ReV. 1989, 89, 1513. (c) Lectka, T. In ActiVe
Metals; Fu¨rstner,A., Ed.; Wiley-VCH: Weinheim, 1995; p 85.
(21) (a) McMurry, J. E.; Lectka, T.; Rico, J. G. Tetrahedron Lett. 1989, 30,
1169. (b) Robertson, G. M. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Ed.; Pergamon Press: Oxford, 1991; p 563.
(22) Fuji, K.; Nakano, S.; Fujita, E. Synthesis 1975, 276.
(26) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999, 1, 953.
(27) Greene, T. W.; Wuts, P. G. M. ProtectiVe Groups in Organic Synthesis,
3rd ed.; John Wiley & Sons, Inc.: New York, 1999; p 41.
J. Org. Chem. Vol. 73, No. 12, 2008 4551

Paquette et al.
SCHEME 9
SCHEME 10
SCHEME 11
significant levels of the ∆9,10 regioisomer. This complication
was bypassed when recourse was alternatively made to the
Peterson olefination.31 Treatment of 31a with freshly prepared
(trimethylsilyl)methyllithium32 followed by p-toluenesulfonic
acid monohydrate-promoted elimination in benzene furnished
32a as the sole product. With pure diene in hand, the time had
come to explore the ring-closing metathesis strategy. High-
dilution conditions were necessary to deter intermolecular
coupling. When proper consideration was accorded to this
parameter, the use of E as catalyst gave 33a in 77% yield.
At this point, hydroboration-oxidation was investigated as
a means for introducing the carbonyl functionality at C13. This
effort quickly made us aware of the sluggish reactivity of the
cyclopentene double bond toward this class of reagents. For
example, no reaction was observed when 33a was heated to 90
°C with 9-borabicyclo[3.3.1]nonane33 or exposed to catecholbo-
rane in the presence of ruthenium(III) chloride.34 Even with
conventional fashion to arrive at 35 (Scheme 10). The hydrobo-
ration of 35 with the BH3 ·THF complex was indeed appreciably
accelerated, reaching completion within 30 min at room
temperature. In addition, the efficiency was significantly im-
proved to above 90%. However, this transformation is poorly
regioselective, with tertiary carbinol 37 being formed competi-
tively with 36. The high proportion of 37 (42%) is attributed to
the directing influence of the unmasked hydroxyl substituent.
This state of affairs, in combination with concerns over
reasonable differentiation of the two OH groups in 36, prompted
the search for an alternative approach to the problem.
Significant advancement came in the form of a three-step
sequence that began with osmium tetraoxide-promoted dihy-
droxylation to give 38 (Scheme 11). Oxidation of this mixture
of diols under Swern conditions resulted in conversion to the
R-ketols 39 without measurable cleavage of the glycolic C-C
bond.36 To our delight, admixture of 39 with samarium(II)
iodide37 proceeded smoothly to generate the cyclopentanone
diastereomers 40 and 41 in 61% overall yield for the three steps.
Our inability to separate 39 into its two constituent isomers
precluded an evaluation of the stereochemistry of the R-ketol
reduction maneuver. The ketone diastereomers 40 and 41 were
amenable to chromatographic purification.
hydroborating reagents as reactive as the borane-tetrahydrofuran
35
and borane-dimethyl sulfide complexes,
neither elevated
temperatures nor prolonged reaction times resulted in the
complete consumption of 33a. Beyond that, the desired product
was isolated as an inseparable mixture of diastereomeric alcohols
in only 36% yield.
We next opted to assess the influence of a free hydroxyl group
on the hydroboration process. For this purpose, 27 was converted
in the predescribed manner to 34, which was desilylated in
(28) (a) Maercker, A. Org React. 1965, 14, 270. (b) Maryanoff, B. E.; Reits,
A. B. Chem. ReV. 1989, 89, 863.
(29) Review Pine, S. H. Org. React. 1993, 43, 1.
(30) Matsubara, S.; Sugihara, M.; Utimoto, K. Synlett 1998, 313, and
references cited therein.
(31) Review: Ager, D. J. Org. React. 1990, 38, 1.
(32) Sommer, L. H.; Mitch, F. A.; Goldberg, G. M. J. Am. Chem. Soc. 1949,
72, 2746.
Of special interest here was not only the 10:1 ratio of 40 to
41 but the obviously greater thermodynamic stability of the 9ꢀ
(33) Scouten, G. G.; Brown, H. C. J. Org. Chem. 1973, 38, 4092.
(34) Review Beletskaya, I.; Pelter, A. Tetrahedron 1997, 4957.
(35) (a) Reviews: Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents;
Academic Press: New York, 1988. (b) Brown, H. C. Boranes in Organic
Chemistry; Cornell University Press: Ithaca, NY, 1972. (c) Brown, H. C. Organic
Syntheses Via Boranes; Wiley: New York, 1975. (d) Cragg, G. M. L.
Organoboranes in Organic Synthesis; Marcel Dekker: New York, 1973.
(36) This was not the case with 2-iodoxybenzoic acid, which was prone to
overoxidize 38 despite claims to the contrary: Frigerio, M.; Santagostino, M.
Tetrahedron Lett. 1994, 35, 8019.
(37) Review: Molander, G. A. Org. React. 1994, 46, 211.
4552 J. Org. Chem. Vol. 73, No. 12, 2008

The Carbohydrate-Sesquiterpene Interface
SCHEME 12
SCHEME 13
incorporated into 42, of allowing for chemoselective reduction
in the presence of the labile lactone functionality. After oxidation
of the coupled product with IBX, an inseparable mixture of
cyclobutanone 43 and its cyclized aldol counterpart 44 was
obtained. Chemical homogeneity was regained when this two-
component system was treated with 10% palladium on carbon
in the presence of triethylsilane.40 These conditions drove the
six-membered ring closure to completion and effectively ac-
complished reduction to the aldehyde level, which materialized
in its enol form as in 45 (74%). Caution was required to bring
about the more advanced reduction of 45 because of the
susceptibility of its lactone ring to cleavage. Ultimately, it was
determined that sodium borohydride in cold (0 °C) methanol
containing potassium dihydrogen phosphate41 was reasonably
effective in delivering diols 46 and 47. Following chromato-
graphic separation, stereochemical assignments were made on
the basis of NOESY measurements. The coformation of 46 and
47 proved not to be problematic as these diastereomers can be
independently transformed into 54 via a comparable route.
Once the hydroxymethyl group in 46 had been selectively
silylated (Scheme 13),42 we set out to functionalize the sterically
congested cyclopentene double bond properly. The deployment
of a hydroboration step again proved ineffective. The process
was neither regioselective nor stereochemically definitive, giving
rise to all four possible isomers. Alternative treatment with
stoichiometric levels of osmium tetroxide followed by cleavage
of the osmate ester with hydrogen sulfide resulted in the almost
exclusive (>95 de) formation of triol 48, which was readily
purified. The outstanding diastereofacial selectivity of this step
may stem from the nearby presence of a tertiary carbinol.43,44
Subsequent to the Swern oxidation of 48, recourse to NOESY
experiments confirmed that the configuration at C9 in 49 was
indeed R. Reductive cleavage of this unneeded C-O bond was
isomer as gauged from our ability to epimerize 41 in methanol
containing K2CO3 to the identical mixture rich in 40. This
observation greatly simplified the throughput from 33a to 40
since the intermediate steps could be usefully carried out on
diastereomeric mixtures without any need for tedious chro-
matographic separation until arrival at 40. The assignment of
stereochemistry to the cyclopentanones was tentatively based
on the method suggested by Kosugi and Uda.6 According to
1
their diagnostic model, the epimer with the larger H NMR
chemical shift difference between the two methylene protons
at C8 possesses a ꢀ-hydrogen at C9. These characteristics are
manifested by 40 and 41 (∆δ ) 0.11 and 0.05, respectively).
Construction of the Unsaturated Lactone Sector and
the End Game for Production of the Natural Enantiomer
of Fomannosin. With construction of the cyclobutane and
cyclopentanone sectors now completed, attention was turned
to generation of the unsaturated lactone ring. We soon became
aware that the necessary step of MOM or MEM deprotection
in more advanced intermediates could not be achieved by any
of the known methods.27 To circumvent this difficulty, recourse
was made to alternative masking of the C4 hydroxyl with the
tert-butyldimethylsilyl group as in 30b (Scheme 9). Our
expectations were that forward progress to 33b would pose no
complications, and this proved to be the case.
(38) (a) Imamoto, T.; Kodera, M.; Yokoyama, M. Bull. Chem. Soc. Jpn. 1982,
55, 2303. (b) Beaulieu, P. L.; Deziel, R.; Brunet, M. L.; Moss, N.; Plante,R.
U.S. cont.-in-part of U.S. 5,574,015, 1998, 24 pp.
(39) Oliver, S. F.; Ho¨genauer, K.; Simic, O.; Antonello, A.; Smith, M. D.;
Ley, S. V. Angew. Chem., Int. Ed. 2003, 42, 5996.
(40) (a) Takayama, H.; Fujiwara, R.; Kasai, Y.; Kitajima, M.; Aimi, N. Org.
Lett. 2003, 5, 2967. (b) Fukuyama, T.; Lin, S.-C.; Li, L. J. Am. Chem. Soc.
1990, 112, 7050.
(41) (a) Fujii, H.; Hirano, N.; Uchiro, H.; Kawamura, K.; Nagase, H. Chem.
Pharm. Bull. 2004, 52, 747. (b) Rabiczko, J.; Urbanczyk-Lipkowska, Z.;
Chemielewski, M. Tetrahedron 2002, 58, 1433.
From this point, the conversion of 33b to diol 42 proceeded
quite smoothly (Scheme 12). We next undertook the regiose-
lective monoesterification of 42 with ethylsulfanylcarbonyl
acetic acid38 as promoted by EDCI.39 The S-ester of monothi-
omalonic acid was selected because it held the prospect, once
(42) Askin, D.; Angst, D.; Danishefsky, S. J. J. Org. Chem. 1987, 52, 622.
(43) Cox, C.; Danishefsky, S. J. Org. Lett. 2001, 3, 2899.
(44) (a) Hodgson, R.; Mahid, T.; Nelson, A. Chem. Commun. 2001, 2076.
(b) Donohoe, T. J.; Fisher, J. W.; Edwards, P. J. Org. Lett. 2004, 6, 465.
J. Org. Chem. Vol. 73, No. 12, 2008 4553

Paquette et al.
SCHEME 14
assignments to C9 on this basis alone. The Kosuki-Uda
correlation6 was viewed to be insufficiently developed to be
universally applicable and error-free. Reversals could materialize
unexpectedly such as in the present examples where the presence
of a C5 hydroxyl and particularly its capacity for intramolecular
hydrogen bonding could induce major conformational changes.
For this reason, we pressed on with both stereoisomeric series.
In the event, 51 and 52 were subjected to a two-step sequence
consisting of initial conversion to their respective cyclic sulfites,
as promoted by triethylamine, followed by DBU-promoted
elimination. Formation of the cyclic sulfites proceeded unevent-
fully. Monitoring of the progress of each of the second reactions
by thin-layer chromatography revealed that a single product,
considered to be 53 and 54, respectively, was being formed
rapidly (5-10 min) in each instance. As time passed, equilibra-
tion of these isomers took place such that similar 1:1 mixtures
of 53 and 54 were in hand after 30 min at rt. ReleVantly, the
proper adjustment of reaction conditions permitted the funneling
1
of all material through 53 in a high state of purity. In the H
NMR spectrum of 53, the chemical shift differences between
methylene C8 protons proved to be unusually small (0.09 ppm).
Direct comparison with the spectral data for 54 showed the ∆δ
in this instance (0.65 ppm) to be somewhat more pronounced
than that for 53. The narrowness of this gap could be a reflection
of intramolecular hydrogen bonding operative between the OH
functionality and C13 carbonyl group in both tricyclic inter-
mediates as suggested upon inspection of molecular models.
The dehydration of the C5 hydroxy group to afford the
cyclobutane double bond proved to be a task as daunting as
expected. No trace amount of cyclobutene product could be
isolated with either Martin sulfurane or Burgess reagent.
Recourse was finally made to formation of the triflate of 53.
Surprisingly, the triflate of 53 was amenable to chromatographic
purification on silica gel. Two chemical events occurred upon
its exposure to DBU in benzene, one being elimination with
introduction of the cyclobutene double bond and the second
being wholesale epimerization to set the C9 configuration as ꢀ.
Although we were initially wary of our ability to control
stereochemistry at C9, the significant thermodynamic advantage
inherent in 55 greatly facilitated matters at this advanced stage
of progress.
next accomplished by exposure of 49 to the action of samarium
iodide in the presence of tert-butyl alcohol as the proton
source.45 The inseparable diastereomers 50 were thereby gener-
ated in 64% yield.
The availability of 50 set the stage for introduction of the
diene unsaturation resident in the target molecule. However,
removal of the PMB group proved not to be as readily
accomplished as expected. When 50 was admixed with DDQ
in a mixture of CH2Cl2 and pH 7 buffer,46 the benzylic position
experienced oxidation to furnish the benzoate.47 The alternative
deprotection procedure involving the use of ceric ammonium
nitrate in aqueous acetonitrile46 resulted in removal of the TBS
group. The desired transformation was ultimately realized with
trifluoroacetic acid under anhydrous conditions.48 Recourse to
this protocol made available a chromatographically separable
two-component mixture consisting of 51 (8%) and 52 (50%)
(Scheme 14). We were not surprised to observe that the chemical
shift difference between the methylene protons at C8 in these
diastereomers are quite distinctive. For 51, the ∆δ value is rather
large (1.12 ppm) when compared to that for 52 (0.18 ppm).
Notwithstanding, we were reluctant to make configurational
The conversion to (+)-fomannosin (1) was successfully
achieved through the use of TBAF in THF at 0 °C. The resulting
primary alcohol, dissolved in purified CDCl3 (stored overnight
with K2CO3 (s) before being filtered through basic alumina prior
to use) to minimize solvent-induced degradation, exhibited a
500 MHz 1H NMR spectrum very closely matching that reported
for fomannosin at lower field strengths.3,4 Synthetic 1 proved
to be dextrorotatory. With these data in hand, it was possible
to make absolute configurational assignments to 51–55 with
confidence.
Total Synthesis of (-)-Fomannosin. En route to (+)-1,
cyclobutanol 22 was produced as the minor constituent (25%)
of the two-component mixture formed upon the zirconocene-
promoted ring contraction of 15. In line with the reactivity of
its diastereomeric counterpart, the hydroxyl group proved
amenable to silylation as in 56, thus allowing for chemoselective
removal of the PMB group and arrival at 57 (Scheme 15). With
this result in hand, we recognized that configurational inversion
at the carbinol center would provide a means for ultimately
delivering (-)-fomannosin. The acquisition of 59 was efficiently
realized by sequential IBX oxidation of 57 followed by reduction
(45) White, J. D.; Somers, T. C. J. Am. Chem. Soc. 1987, 109, 4424.
(46) Wang, Y.; Babirad, S. A.; Kishi, Y. J. Org. Chem. 1992, 57, 468.
(47) Blakemore, P. R.; Kim, S.-K.; Schulze, V. K.; White, J. D.; Yokochi,
A. F. T. J. Chem. Soc., Perkin Trans. 1 2001, 1831.
(48) (a) Yan, L.; Kahne, D. Synlett 1995, 523. (b) Bonzide, A.; Sauve´, G.
Tetrahedron Lett. 1999, 40, 2883.
4554 J. Org. Chem. Vol. 73, No. 12, 2008

The Carbohydrate-Sesquiterpene Interface
SCHEME 15
SCHEME 16
with sodium borohydride. The preponderance of 59, a likely
reflection of steric control operating in 58, was 10:1. The
subsequent critical conversion of 59 into ent-30b was realized
uneventfully, thereby affording a six-step bridge suitable for
the production of unnatural fomannosin.
Experimental Section
(-)-(3aS,5S,6R,6aS)-Methyl Dihydro-5-tert-butyldiphenylsi-
loxymethyl)-2,2-dimethyl-5H-furo[3,2-d][1,3]dioxole-6-carboxy-
late (17). To a THF solution (10 mL) of diisopropylamine (1.40
mL, 10 mmol) cooled to -30 °C was added n-butyllithium (1.5 M
in hexanes, 3.33 mL, 5.0 mmol), and the resulting solution was
stirred for 20 min in the cold before proceeding to -78 °C. Ester
10 (1.01 g, 5.00 mmol) in 6 mL of THF was quickly introduced,
and 1 h was allowed to elapse before a THF solution (15 mL) of
monomeric formaldehyde was added. The resulting mixture was
warmed to -10 °C, quenched with saturated NH4Cl solution (1
mL), and partitioned between ethyl acetate (300 mL) and water
(50 mL). The separated aqueous phase was evaporated, the residue
was dissolved in 1:1 methanol/ethyl acetate, and this solution was
filtered through Celite. After drying and evaporation of the filtrate,
the residue was subjected to MPLC on silica gel (elution with 2:3
hexanes/ethyl acetate) to give the alcohol (550 mg, 52% based on
Ultimately, ent-30b was converted to levorotatory fomannosin
via the synthetic protocol developed earlier for (+)-1 (Scheme
16). The routes to these antipodes feature highly efficient
functionalization reactions of extensively substituted cyclobutanes.
In summary, we have achieved asymmetric syntheses of (+)-
and (-)-fomannosin from a common carbohydrate precursor.
The key steps involved the initial elaboration of 21 from R-D-
glucose, assembly of the cyclopentene subunit by ring-closing
metathesis, and construction of the unsaturated lactone sector
in advance of functional group manipulation. Atoms C4 to C9
of the fomannosin backbone arise from the glucose starting
material as a result of zirconocene-promoted deoxygenative ring
contraction. A great deal of stereochemical control seems
available and it should prove feasible to modify configuration
at C7 and C9 appropriately as desired for assessing structure/
activity relationships.
recovered 10) as a white solid: mp 103-104 °C; IR (film, cm-1
)
3499, 1751, 1257; 1H NMR (300 MHz, CDCl3) δ 5.84 (d, J ) 3.4
Hz, 1 H), 4.70 (t, J ) 4.0 Hz, 1 H), 3.80-3.71 (m, 1 H), 3.74 (s,
3 H), 3.54 (dd, J ) 7.2, 11.6 Hz, 1 H), 2.65 (d, J ) 14.2 Hz, 1 H),
2.56 (t, J ) 6.7 Hz, 1 H), 2.19 (dd, J ) 4.7, 14.2 Hz, 1 H), 1.42
J. Org. Chem. Vol. 73, No. 12, 2008 4555

Paquette et al.
(s, 3 H), 1.26 (s, 3 H); 13C NMR (75 MHz, CDCl3) δ 172.6, 112.6,
106.5, 88.2, 80.2, 66.1, 52.4, 35.8, 25.6, 25.5; HRMS (ES) m/z (M
70.5, 54.6, 39.6, 26.9, 19.5; HRMS (ES) m/z (M + Na)+ calcd
435.1962, obsd 435.1961; [R]22 -21.5 (c 1.05, CHCl3).
D
+ Na)+ calcd 255.0839, obsd 255.0861; [R]22 -66.6 (c 1.49,
For 19r: IR (film, cm-1) 3558, 1428; 1H NMR (300 MHz,
CDCl3) δ 7.69-7.64 (m, 4 H), 7.46-7.35 (m, 6 H), 5.93 (dd, J )
10.8, 17.4 Hz, 1 H), 5.26 (dd, J ) 1.4, 17.3 Hz, 1 H), 5.05 (dd, J
) 1.4, 10.8 Hz, 1 H), 4.89 (d, J ) 4.6 Hz, 1 H), 4.52-4.45 (m, 1
H), 3.50 (d, J ) 10.4 Hz, 1 H), 3.49 (s, 3 H), 3.44 (d, J ) 10.4 Hz,
1 H), 2.58 (dd, J ) 8.1, 12.1 Hz, 1 H) 1.78 (dd, J ) 9.4, 12.1 Hz,
1 H), 1.06 (s, 9 H); 13C NMR (75 MHz, CDCl3) δ 141.1, 135.8,
135.7, 133.3, 129.9, 129.8, 127.8, 113.6, 103.1, 85.6, 72.7, 69.4,
55.3, 39.4, 27.0, 19.4; HRMS (ES) m/z (M + Na)+ calcd 435.1962,
D
CHCl3).
To a DMF solution (15 mL) of the above alcohol (900 mg, 3.88
mmol) and imidazole (1.32 g, 19.4 mmol) was added tert-
butyldiphenylsilyl chloride (1.28 g, 4.66 mmol), and the reaction
mixture was stirred overnight at rt before being quenched with water
(100 mL) and extracted with ethyl acetate (3 - 100 mL). After
drying and evaporation of the combined organic layers, the product
was chromatographed on silica gel (elution with 7:1 hexanes/ethyl
acetate) to furnish 17 (1.64 g, 90%) as a colorless oil: IR (film,
cm-1) 1766, 1732. 1428; 1H NMR (300 MHz, CDCl3) δ 7.68-7.64
(m, 4 H), 7.45-7.35 (m, 6 H), 5.90 (d, J ) 3.5 Hz, 1 H), 4.74 (dd,
J ) 3.5, 4.9 Hz, 1 H), 3.85 (d, J ) 10.4 Hz, 1 H), 3.74 (s, 3 H),
3.69 (d, J ) 10.4 Hz, 1 H), 2.72 (d, J ) 14.2 Hz, 1 H), 2.29 (dd,
J ) 4.9, 14.2 Hz, 1 H), 1.48 (s, 3 H), 1.31 (s, 3 H), 1.03 (s, 9 H);
13C NMR (75 MHz, CDCl3) 172.7, 135.6 (2 C), 132.9, 132.7, 129.8,
127.8, 112.5, 106.8, 88.6, 80.3, 67.6, 52.3, 36.5, 26.6, 26.0, 25.9,
19.2; HRMS (ES) m/z (M + Na)+ calcd 493.2017, obsd 493.2027;
obsd 435.1966; [R]22 +78.4 (c 0.84, CHCl3).
D
O-Benzylation of 19ꢀ. Sodium hydride (1.15 g of 60% mineral
oil dispersion, 28.6 mmol) was added in one portion into a solution
of 19ꢀ (9.85 g, 23.9 mmol) in 220 mL of DMF at 0 °C. After 2
min, a solution of p-methoxybenzyl bromide (5.28 g, 26.3 mmol)
in 30 mL of DMF was introduced dropwise, and stirring was
maintained at rt for 4 h. After quenching with saturated NaHCO3
solution and dilution with ethyl acetate and water, the separated
aqueous phase was extracted with ethyl acetate, and the combined
organic layers were washed with brine, dried, and concentrated to
afford 15 (10.85 g, 82%) as a colorless oil spectroscopically
identical to the material described in the Supporting Information.
O-Benzylation of 19r. Sodium hydride (1.08 g of 60% mineral
oil dispersion, 26.9 mmol) was added in one portion into a solution
of 19r (8.54 g, 20.7 mmol) in 200 mL of DMF at 0 °C. After 2
min, a solution of p-methoxybenzyl bromide (4.99 g, 24.8 mmol)
in 25 mL of DMF was added dropwise, and stirring was maintained
at rt for 1 h. Following the introduction of 0.2 equiv each of
additional sodium hydride and the bromide, agitation was prolonged
for 2 h and the predescribed workup protocol was followed. There
[R]22 -28.2 (c 1.43, CHCl3).
D
(+)-((3aS,5S,6R,6aS)-Dihydro-2,2-dimethyl-6-vinyl-5H-furo[3,2-
d][1,3]dioxol-5-yl)methanol tert-Butyldiphenylsilyl Ether (18). To
a CH2Cl2 solution of 17 (905 mg, 1.92 mmol) cooled to -78 °C
was added diisobutylaluminum hydride (4.8 mL of 1.0 M in
hexanes). The reaction mixture was stirred at this temperature for
2 h prior to being quenched with methanol (1 mL) and 20% sodium
potassium tartrate solution (20 mL). Stirring was maintained until
clear phase separation was achieved. The aqueous layer was
extracted with CH2Cl2 (3 × 00 mL), and the combined organic
phases were washed with brine, dried, and evaporated to leave the
aldehyde that was used directly without purification.
was isolated 9.06 g (82%) of 20 as a colorless oil: IR (film, cm-1
)
To a solution of methyltriphenylphosphonium bromide (2.13 g,
5.97 mmol) in THF (15 mL) was added n-butyllithium (3.3 mL of
1.5 M in hexanes) at -30 °C, and stirring was maintained for 20
min before the above aldehyde dissolved in THF (10 mL) was
introduced. The reaction mixture was allowed to warm to rt, stirred
for 3 h, and quenched with saturated NaHCO3 solution (5 mL).
Dilution with ethyl acetate (300 mL) and washing with brine
followed. The organic phase was dried and freed of solvent, and
the residue was purified by chromatography on silica gel (elution
with 15:1 hexanes/ethyl acetate) to provide 682 mg (81% over two
steps) of 18 as a colorless oil: IR (film, cm-1) 1428, 1213, 1113;
1H NMR (300 MHz, CDCl3) δ 7.70-7.64 (m, 4 H), 7.46-7.35
(m, 6 H), 5.99 (d, J ) 3.9 Hz, 1 H), 5.95 (dd, J ) 10.9, 17.4 Hz,
1 H), 5.33 (dd, J ) 1.3, 17.4 Hz, 1 H), 5.10 (dd, J ) 1.3, 10.9 Hz,
1 H), 4.90-4.86 (m, 1 H), 3.58 (d, J ) 10.6 Hz, 1 H), 3.37 (d, J
) 10.6 Hz, 1 H), 2.54 (dd, J ) 6.5, 13.8 Hz, 1 H), 2.14 (dd, J )
1.4, 13.8 Hz, 1 H), 1.51 (s, 3 H), 1.34 (s, 3 H), 1.05 (s, 9 H); 13C
NMR (75 MHz, CDCl3) δ 139.6, 135.6 (2 C), 133.2, 132.9, 129.7
(2 C), 127.7, 114.8, 112.5, 106.7, 88.6, 82.0, 69.4, 38.0, 26.9, 26.8,
19.2; HRMS (ES) m/z (M + Na)+ calcd 461.2119, obsd 461.2123;
[R]22D +0.40 (c 1.97, CHCl3). Anal. Calcd for C26H34O4Si: C, 71.19;
H, 7.81. Found: C, 71.30; H, 7.82.
1
1613, 1512; H NMR (300 MHz, CDCl3) δ 7.65-7.61 (m, 4 H),
7.45-7.34 (m, 6 H), 7.27-7.25 (m, 2 H), 6.87-6.82 (m, 2 H),
5.91 (dd, J ) 10.9, 17.4 Hz, 1 H), 5.21 (dd, J ) 1.3, 17.4 Hz, 1
H), 5.03 (dd, J ) 1.3, 10.9 Hz, 1 H), 4.80 (d, J ) 4.3 Hz, 1 H),
4.52 (m, 2 H), 4.26-4.18 (m, 1 H), 3.79 (s, 3 H), 3.50-3.41 (m,
5 H), 2.43 (dd, J ) 8.1, 12.7 Hz, 1 H), 2.01 (dd, J ) 10.5, 11.7
Hz, 1 H), 1.01 (s, 9 H); 13C NMR (75 MHz, CDCl3) δ 159.4, 141.1,
135.8, 135.7, 133.4, 133.2, 130.2, 129.84, 129.77, 129.75, 127.8,
113.9, 113.7, 102.3, 84.6, 78.5, 72.2, 69.4, 55.4, 35.9, 26.9, 19.4;
HRMS (ES) m/z (M + Na)+ calcd 555.2537, obsd 555.2516; [R]22
D
+59.5 (c 0.80, CHCl3).
(-)-(1S,2R,4R)-4-(4-Methoxybenzyloxy)-2-tert-butyldiphenylsi-
loxymethyl)-2-vinylcyclobutanol (21) and (+)-[1R,2R,4R)-4-(4-
Methoxybenzyloxy)-2-tert-butyldiphenylsiloxymethyl)-2-vinylcy-
clobutanol (22). A solution of zirconocene dichloride (2.68 g, 9.17
mmol) in THF (150 mL) was treated with a solution of n-
butyllithium in hexanes (1.45 M, 11.0 mL, 15.9 mmol) at -78 °C.
Stirring was maintained at this temperature for 1 h before a solution
of 15 (3.25 g, 6.11 mmol) in THF (25 mL) was introduced. The
resulting solution was allowed to warm slowly, stirred overnight
at rt, and passed through a short silica gel column with ether. The
eluate was concentrated and purified by flash chromatography on
silica gel (elution with hexanes/ethyl acetate 5:1) to give 21 (1.85
g, 60%) and 22 (0.75 g, 25%) as colorless oils.
Acid Hydrolysis of 18. To a solution of 18 (370 mg, 12.2 mmol)
in methanol (92 mL) was added a solution of concentrated HCl
(2.0 mL) in methanol (8 mL) at rt. The reaction mixture was stirred
for 1.5 h, quenched with solid NaHCO3, and freed of methanol.
The residue was dissolved in CH2Cl2, dried, and concentrated. Flash
chromatography of the residue on silica gel (elution with 6:1 to
5:1 hexanes/ethyl acetate) afforded 151 mg (43%) of the ꢀ-isomer
and 106 mg (30%) of the R-isomer.
For 21: IR (film, cm-1) 1612, 1514, 1428; 1H NMR (300 MHz,
CDCl3) δ 7.67-7.63 (m, 4 H), 7.47-7.36 (m, 6 H), 7.30-7.25
(m, 2 H), 6.91-6.86 (m, 2 H), 5.95 (dd, J ) 17.6, 10.9 Hz, 1 H),
5.26 (dd, J ) 10.9, 1.3 Hz, 1 H), 5.09 (dd, J ) 17.6, 1.3 Hz, 1 H),
4.46 (s, 2 H), 4.46-4.40 (m, 1 H), 4.20-4.14 (m, 1 H), 3.81 (s, 3
H), 3.64 (d, J ) 10.1 Hz, 1 H), 3.46 (d, J ) 10.1 Hz, 1 H), 2.57
(br s, 1 H), 2.27 (ddd, J ) 12.7, 6.8, 2.3 Hz, 1 H), 2.10 (dd, J )
12.7, 4.9 Hz, 1 H), 1.06 (s, 9 H); 13C NMR (75 MHz, CDCl3) δ
159.3, 137.6, 135.6 (4 C), 133.3, 133.2, 129.9, 129.7 (2 C), 127.7
(4 C), 116.5, 113.8 (2 C), 71.6, 71.3, 70.8, 67.9, 55.2, 49.9, 31.1,
26.8 (3 C), 19.3; HRMS (ES) m/z (M + Na)+ calcd 525.2432, obsd
For 19ꢀ: IR (film, cm-1) 3438, 1472, 1428; 1H NMR (300 MHz,
CDCl3) δ 7.68-7.65 (m, 4 H), 7.46-7.34 (m, 6 H), 6.20 (dd, J )
17.8, 17.3 Hz, 1 H), 5.48 (dd, J ) 1.6, 17.2 Hz, 1 H), 5.18 (dd, J
) 1.6, 10.7 Hz, 1 H), 4.94 (s, 1 H), 4.16-4.09 (m, 1 H), 3.60 (s,
2 H), 3.28 (s, 3 H), 2.34 (dd, J ) 5.2, 13.7 Hz, 1 H), 1.95 (d, J )
13.6 Hz,1 H), 1.09 (s, 9 H); 13C NMR (75 MHz, CDCl3) δ 143.0,
135.78, 135.75, 133.5, 133.4, 129.8, 127.7, 113.5, 110.1, 87.0, 76.9,
525.2429; [R]24 -25.6 (c 1.28, CHCl3).
D
4556 J. Org. Chem. Vol. 73, No. 12, 2008

The Carbohydrate-Sesquiterpene Interface
For 22: IR (film, cm-1) 1613, 1587, 1513, 1428; 1H NMR (300
MHz, CDCl3) δ 7.68-7.64 (m, 4 H), 7.47-7.35 (m, 6 H),
7.32-7.27 (m, 2 H), 6.91-6.86 (m, 2 H), 5.89 (dd, J ) 17.7, 11.0
Hz, 1 H), 5.30 (dd, J ) 11.0, 1.1 Hz, 1 H), 5.19 (dd, J ) 17.7, 1.1
Hz, 1 H), 4.49 (s, 2 H), 4.22 (d, J ) 6.4 Hz, 1 H), 3.81 (s, 3 H),
3.77 (dt, J ) 8.4, 6.4 Hz, 1 H), 3.65 (d, J ) 10.2 Hz, 1 H), 3.50
(d, J ) 10.2 Hz, 1 H), 2.13 (dd, J ) 11.2, 8.4 Hz, 1 H), 1.79 (dd,
J ) 11.2, 8.4 Hz, 1 H), 1.72 (br s, 1 H), 1.09 (s, 9 H); 13C NMR
(75 MHz, CDCl3) δ 159.1, 137.0, 135.6 (4 C), 133.4 (2 C), 130.4,
129.7 (2 C), 129.4 (2 C), 127.7 (4 C), 117.1, 113.7 (2 C), 78.0,
75.1, 70.5, 67.3, 55.2, 45.4, 27.7, 26.9 (3 C), 19.4; HRMS (ES)
romethane (5:1:1) to afford a slightly yellow solid, which was used
directly for the next step.
The above triflate was taken up in 0.2 mL of benzene. DBU
(1.7 µL in 30 µL of benzene, 11.4 µmol) was added at rt. After 2 h
at rt, the mixture was loaded onto a silica gel column, eluted with
hexanes/ethyl acetate/dichloromethane (6:1:1 with 0.5% triethy-
lamine) to afford 55 (less polar, 0.5 mg, 35% over 2steps) and its
C-9 epimer (more polar, 0.3 mg, 21%) both as colorless oils;
For 55: 1H NMR (500 MHz, C6D6) δ 6.71 (d, J ) 2.2 Hz, 1 H),
6.16 (d, J ) 2.3 Hz, 1 H), 5.02 (d, J ) 10.0 Hz, 1 H), 4.86 (d, J
) 15.3 Hz, 1 H), 4.59 (d, J ) 15.4 Hz, 1 H), 3.97 (d, J ) 10.1 Hz,
1 H), 3.18 (dd, J ) 9.4, 11.4 Hz, 1 H), 1.93 (d, J ) 17.8 Hz, 1 H),
1.68 (d, J ) 18.1 Hz, 1H), 1.53-1.49 (m, 2H), 1.03 (s, 9H), 0.83
(s, 3H), 0.63 (s, 3H), 0.13 (s, 3H), 0.12 (s, 3H); HRMS (ES) m/z
(M + Na)+ calcd 399.1968, obsd 399.1986.
(+)-Fomannosin (1). Tetrabutylammonium fluoride (4.0 µL, 1.0
M in THF, 4.0 µmol) was added dropwise to a solution of 55 (0.5
mg, 1.33 µmol) in tetrahydrofuran (0.5 mL) at 0 °C under argon.
After 20 min at 0 °C, the reaction mixture was filtered through a
short pad of silica gel, washing with ethyl acetate/dichloromethane
(2/1, with 0.5% triethylamine). The filtrate was concentrated. The
residue was purified by preparative thin layer chromatography (silica
gel, ethyl acetate/dichloromethane 1/1 with 0.5% triethylamine) to
afford (+)-1 (0.3 mg, 86%) as a colorless oil; IR (film, cm-1) 3580,
1724, 1709, 1461, 1408; 1H NMR (500 MHz, CDCl3) δ 6.89 (d, J
) 2.4 Hz, H-6, 1 H), 6.69 (d, J ) 2.4 Hz, H-5, 1 H), 4.90 (d, J )
10.2 Hz, H-8, 1 H), 4.42 (dd, J ) 5.2, 13.8 Hz, H-1, 1 H), 4.35
(dd, J ) 6.0, 13.8 Hz, H-1, 1 H), 4.28 (d, J ) 10.2 Hz, H-8, 1 H),
3.18 (dd, J ) 8.9, 12.4 Hz, H-9, 1 H), 2.37-2.35 (m, OH, 1H),
2.22 (d, J ) 18.5 Hz, H-12, 1 H), 1.95 (d, J ) 18.5 Hz, H-12, 1
H), 1.73 (ddd, J ) 2.4, 8.7, 12.6 Hz, H-10, 1 H), 1.57 (H-10, 1 H,
overlap with H2O peak), 1.16 (s, 3 H), 1.10 (s, 3 H); 13C NMR
(150 MHz, CD2Cl2) δ 218.37 (C-13), 165.72 (C-3), 154.85 (C-4),
147.96 (C-5), 139.53 (C-6), 73.50 (C-8), 58.52 (C-1), 53.61 (C-
12), 52.79 (C-7), 46.60 (C-9), 38.39 (C-10), 32.94 (C-11), 29.65
(CH3), 27.89 (CH3); HRMS (ES) m/z (M + Na)+ calcd 285.1103,
obsd 285.1119; [R]22D + 160 (c 0.02, CHCl3). See the Supporting
Information for a comparison of 1H and 13C NMR data for synthetic
and natural fomannosin. Due to small quantities of material, 13C
data were extrapolated from HSQC and HMBC data.
m/z (M + Na)+ calcd 525.2432, obsd 525.2423; [R]24 +10.2 (c
D
0.66, CHCl3).
(+)-(1S,7R)-5-((tert-Butyldimethylsilyloxy)methyl)-1-((S)-4,4-di-
methyl-2-oxocyclopentyl)-7-hydroxy-3-oxabicyclo[4.2.0]oct-5-en-
4-one (53) and (+)-(1S,7R)-5-((tert-Butyldimethylsilyloxy)
methyl)-1-((R)-4,4-dimethyl-2-oxocyclopentyl)-7-hydroxy-3-
oxabicyclo[4.2.0]oct-5-en-4-one (54). Thionyl chloride (1.08 µL,
0.015 mmol) was added slowly to a solution of 52 (5.1 mg, 0.012
mmol) and triethylamine (5.17 µL, 0.037 mmol) in 0.6 mL of dry
CH2Cl2 at 0 °C under argon. After 20 min at this temperature, the
solution was poured into a mixture of ethyl acetate, water, and a
small amount of triethylamine. The organic layer was washed with
brine, dried, and concentrated. The solid residue was immediately
taken up in 1.0 mL of dry CH2Cl2 and cooled to 0 °C. DBU (5.6
µL, 0.037 mmol) was added at 0 °C under argon. After 30 min at
0 °C, the mixture was loaded directly onto a silica gel column,
eluted with hexanes/ethyl acetate (3:1, with 1% ethanol) to afford
53 (1.7 mg, 35%) and 54 (1.5 mg, 31%), both as white solids.
For 53: IR (film, cm-1) 3456, 1725, 1463; 1H NMR (500 MHz,
CDCl3) δ 5.09 (dt, J ) 2.4, 6.7 Hz, 1 H), 4.57 (d, J ) 14.6 Hz, 1
H), 4.39 (d, J ) 14.6 MHz, 1 H), 4.24 (d, J ) 10.8 Hz, 1 H), 4.25
(d, J ) 10.8 Hz, 1 H), 3.38 (d, J ) 7.3 Hz, 1 H), 3.33 (dd, J ) 8.3,
13.1 Hz, 1 H), 2.35 (dd, J ) 6.6, 13.4 Hz, 1 H), 2.31 (d, J ) 20.1
Hz, 1 H), 2.15 (dd, J ) 2.7, 13.4 Hz, 1 H), 2.08 (d, J ) 19.0 Hz,
1 H), 2.06-2.02 (m, 1 H), 1.71 (t, J ) 12.8 Hz, 1 H), 1.22 (s, 3
H), 1.12 (s, 3 H), 0.92 (s, 9 H), 0.13 (s, 3 H), 0.12 (s, 3 H); 13C
NMR (125 MHz, CDCl3) δ 221.1, 163.5, 125.2, 76.0, 70.6, 59.8,
54.4, 53.3, 44.8, 40.1, 38.5, 33.4, 30.1, 27.9, 26.1 (3 C), 18.5, -5.3,
-5.4; HRMS (ES) m/z (M + Na)+ calcd 417.2073, obsd 417.2089;
(-)-tert-Butyl(((1S,2R,3R)-2-(tert-butyldimethylsilyloxy)-3-(4-
methoxybenzyloxy)-1-vinylcyclobutyl)methoxy)diphenylsilane (56).
To a solution of 22 (1.40 g, 2.78 mmol) and imidazole (0.95 g,
13.9 mmol) in CH2Cl2 (10 mL) was added TBSCl (0.63 g, 4.18
mmol) at rt. The mixture was stirred at rt for 3 h before being
quenched with MeOH (3 mL). The resulting solution was stirred
at rt for 2 h, diluted with CH2Cl2 (100 mL), washed with saturated
NaHCO3 solution (1×) and brine (1×), dried, and concentrated.
The residue was purified by flash chromatography on silica gel
(elution with hexanes/ethyl acetate, 30/1) to afford 56 as a pale
[R]22 +22.4 (c 0.36, CHCl3).
D
For 54: IR (film, cm-1) 3410, 1728, 1694; 1H NMR (500 MHz,
CDCl3) δ 5.13 (d, J ) 6.3 Hz, 1 H), 4.83 (d, J ) 10.5 Hz, 1 H),
4.65 (dd, J ) 1.0, 15.8 Hz, 1 H), 4.37 (d, J ) 15.8 Hz, 1 H), 4.17
(d, J ) 10.5 Hz, 1 H), 3.12 (dd, J ) 8.5, 12.6 Hz, 1 H), 2.34 (dd,
J ) 6.7, 13.5 Hz, 1 H), 2.29 (d, J ) 2.2 Hz, 1 H), 2.22-2.15 (m,
2 H), 2.07 (d, J ) 18.3 Hz, 1 H), 2.02 (dd, J ) 2.6, 13.4 Hz, 1 H),
1.92 (ddd, J ) 2.4, 8.4, 12.8 Hz, 1 H), 1.25 (s, 3 H), 1.08 (s, 3 H),
0.93 (s, 9 H), 0.13 (s, 6 H); 13C NMR (125 MHz, CDCl3) δ 218.6,
163.7, 156.9, 125.7, 76.3, 69.9, 61.1, 54.2, 49.1, 44.3, 40.1, 35.8,
33.7, 30.0, 27.8, 26.1 (3 C), 18.5, -5.3, -5.4; HRMS (ES) m/z
(M + Na)+ calcd 417.2073, obsd 417.2087; [R]22D +100.8 (c 0.19,
CHCl3).
1
yellow oil (1.57 g, 91%): IR (film, cm-1) 1613, 1514, 1249; H
NMR (300 MHz, CDCl3) δ 7.69-7.63 (m, 4 H), 7.46-7.34 (m, 6
H), 7.27 (d, J ) 8.7 Hz, 2 H), 6.88 (d, J ) 8.7 Hz, 2 H), 5.95 (dd,
J ) 17.7, 10.9 Hz, 1 H), 5.15 (dd, J ) 10.9, 1.4 Hz, 1 H), 5.05
(dd, J ) 17.7, 1.4 Hz, 1 H), 4.45 (ABq, J ) 11.5 Hz, ∆ν) 18.6
Hz, 2 H), 4.38 (d, J ) 6.6 Hz, 1 H), 3.82 (s, 3 H), 3.82-3.74 (m,
1 H), 3.56 (d, J ) 10.3 Hz, 1 H), 3.39 (d, J ) 10.3 Hz, 1 H), 2.03
(dd, J ) 10.8, 8.1 Hz, 1 H), 1.77 (dd, J ) 10.8, 8.6 Hz, 1 H), 1.10
(s, 9 H), 0.90 (s, 9 H), 0.10 (s, 3 H), 0.05 (s, 3 H); 13C NMR (75
MHz, CDCl3) δ 159.0, 138.0, 135.7, 133.6, 133.5, 130.9, 129.6,
129.1, 127.6, 115.2, 113.7, 77.9, 74.2, 70.3, 66.5, 55.3, 45.3, 27.7,
26.9, 25.8, 19.4, 18.0, -4.7; HRMS (ES) m/z (M + Na)+ calcd
Equilibration of 53 and 54. A solution of compound 54 (2.6
mg, 6.6 µmol) in dichloromethane (0.6 mL) was treated with DBU
(3.0 µL, 19.8 µmol) at rt. After 45 min at rt, the mixture was loaded
directly onto a silica gel column and eluted with hexanes/ethyl
acetate (3.5:1, with 1% ethanol) to afford 53 (1.4 mg, 54%) and
54 (1.2 mg, 46%).
(S)-5-((tert-Butyldimethylsilyloxy)methyl)-1-((S)-4,4-dimethyl-2-
oxocyclopentyl)-3-oxabicyclo[4.2.0]octa-5,7-dien-4-one (55). Triflic
anhydride (6.4 µL, 0.038 mmol) was added slowly to a solution of
53 (1.5 mg, 3.80 µmol) and pyridine (6.1 µL, 0.076 mmol) in dry
dichloromethane (0.8 mL) at 0 °C under argon. After 15 min at 0
°C, isopropyl alcohol (2.9 µL, 0.038 mmol) was added. After 20
min at 0 °C, DBU (5.7 µL, 0.028 mmol) was introduced dropwise
at 0 °C. After another 5 min, the reaction mixture was loaded onto
a silica gel column, eluted with hexanes/ethyl acetate/dichlo-
639.3296, obsd 639.3247; [R]24 -18.6 (c 1.7, CHCl3).
D
(-)-(1R,2R,3S)-2-(tert-Butyldimethylsilyloxy)-3-((tert-butyldiphe-
nylsilyloxy)methyl)-3-vinylcyclobutanol (57). Trifluoroacetic acid
(20 mL) was added quickly to a solution of 56 (21.54 g, 34.91
mmol) in 200 mL of dry CH2Cl2 at rt under argon. After 20 min,
the reaction mixture was cooled to 0 °C and quenched with an
excess amount (50 mL) of triethylamine. The solution was
J. Org. Chem. Vol. 73, No. 12, 2008 4557

Paquette et al.
concentrated. The residue was filtered through a short pad of silica
gel, and the pad was washed with hexanes/ethyl acetate (10/1). The
filtrate was concentrated, and the residue was purified by flash
chromatography (silica gel, elution with hexanes/ethylacetate 15/
(0.240 g, 0.483 mmol) and PMB imidate (0.341 g, 1.208 mmol) in
10 mL of dry ether at rt under argon. The progress of reaction was
closely monitored by TLC. Additional TfOH (0.13 µL, 1.45 µmol)
in 50 µL of dry ether was added every 30 min until the starting
material had been consumed. The reaction mixture was quenched
with excess triethylamine and concentrated. The residue was purified
by flash chromatography (silica gel, elution with hexanes/ethyl
acetate 25/1) to afford the protected alkene as a colorless oil (0.191
1) to afford 57 as a colorless oil (15.58 g, 90%): IR (neat, cm-1
)
1
3422, 1472, 1428; H NMR (500 MHz, CDCl3) δ 7.72-7.68 (m,
4 H), 7.48-7.40 (m, 6 H), 5.97 (dd, J ) 10.9, 17.7 Hz, 1 H), 5.19
(dd, J ) 1.2, 10.9 Hz, 1 H), 5.11 (dd, J ) 1.4, 17.7 Hz, 1 H), 4.26
(d, J ) 6.3 Hz, 1 H), 3.99-3.94 (m, 1 H), 3.59 (d, J ) 10.4 Hz,
1 H), 3.41 (d, J ) 10.4 Hz, 1 H), 2.18 (dd, J ) 8.5, 10.9 Hz, 1 H),
1.79 (dd, J ) 8.7, 10.9 Hz, 2 H), 1.14 (s, 9 H), 0.95 (s, 9 H), 0.14
(s, 3 H), 0.11 (s, 3 H); 13C NMR (125 MHz, CDCl3) δ 137.8, 135.8,
133.62, 133.55, 129.8, 127.78, 127.77, 115.4, 76.3, 72.5, 66.2, 44.9,
30.0, 27.1, 26.0, 19.5, 18.2, -4.4, -4.7; HRMS m/z (M + Na)+
1
g, 64%): IR (neat, cm-1) 1513, 1428, 1248; H NMR (500 MHz,
CDCl3) δ 7.72-7.68 (m, 4 H), 7.49-7.41 (m, 6 H), 7.32 (d, J )
8.6 Hz, 1 H), 6.91 (d, J ) 8.6 Hz, 1 H), 6.22 (dd, J ) 11.0, 17.8
Hz, 1 H), 5.16 (dd, J ) 1.4, 11.0 Hz, 1 H), 4.99 (dd, J ) 1.4, 17.8
Hz, 1 H), 4.62 (d, J ) 11.6 Hz, 1 H), 4.57 (dd, J ) 1.8, 5.3 Hz, 1
H), 4.42 (d, J ) 11.6 Hz, 1 H), 4.13-4.09 (m, 1 H), 3.84 (s, 3 H),
3.66 (d, J ) 10.3 Hz, 1 H), 3.54 (d, J ) 10.3 Hz, 1 H), 2.26 (ddd,
J ) 2.3, 6.9, 9.9 Hz, 1 H), 2.07 (dd, J ) 4.8, 12.1 Hz, 1 H), 1.12
(s, 9 H), 0.96 (s, 9 H), 0.11 (s, 3 H), 0.09 (s, 3 H); 13C NMR (125
MHz, CDCl3) δ 159.0, 139.3, 135.75, 135.72, 133.47, 133.45,
131.1, 129.7, 129.2, 127.71, 127.70, 114.2, 113.7, 72.9, 72.1, 70.5,
67.1, 55.3, 49.7, 32.0, 31.9, 29.74, 29.70, 29.4, 27.0, 26.0, 22.7,
19.4, 18.4, 14.2, -4.6, -4.7; HRMS m/z (M + Na)+ calcd
calcd 519.2727, obsd 519.2707; [R]22 -11.6 (c 1.5, CHCl3).
D
(+)-(2R,3S)-2-(tert-Butyldimethylsilyloxy)-3-((tert-butyldiphenyl-
silyloxy)methyl)-3-vinylcyclobutanone (58). A solution of 57 (1.360
g, 2.738 mmol) and IBX (2.300 g, 8.213 mmol) in 30 mL of DMSO
was stirred at rt for 4 h. The reaction mixture was diluted with
ether and H2O. The organic layer was washed twice with brine,
dried, and filtered. The filtrate was concentrated. The residue was
purified by flash chromatography (elution with silica gel, hexanes/
ethyl acetate 25/1) to afford 58 (1.265 g, 93%) as colorless oil: IR
639.3302, obsd 639.3310; [R]22 +0.81 (c 1.1, CHCl3). The
D
spectrometric data are in accordance with those of the precursor of
30b.
1
(neat, cm-1) 1791, 1472, 1428; H NMR (500 MHz, CDCl3) δ
7.67-7.62 (m, 4 H), 7.47-7.38 (m, 6 H), 5.86 (dd, J ) 11.0, 17.7
Hz, 1 H), 5.23 (dd, J ) 0.5, 11.1 Hz, 1 H), 5.16 (t, J ) 2.6 Hz, 1
H), 5.09 (dd, J ) 0.8, 17.7 Hz, 1 H), 3.86 (d, J ) 10.3 Hz, 1 H),
3.71 (d, J ) 10.3 Hz, 1 H), 3.08 (dd, J ) 3.0, 16.2 Hz, 1 H), 2.49
(dd, J ) 2.2, 16.2 Hz, 1 H), 1.10 (s, 9 H), 0.91 (s, 9 H), 0.13 (s,
3 H), 0.08 (s, 3 H); 13C NMR (125 MHz, CDCl3) δ 204.7, 135.74,
135.71, 135.1, 133.1, 133.0, 130.0, 128.0, 127.9, 117.2, 65.1, 43.9,
43.7, 27.1, 25.8, 19.5, 18.3, -4.5, -4.9; HRMS m/z (M + Na)+
Five drops of a 0.1% solution of Sudan III in dichloromethane
were added to a solution of the above product (11.80 g, 19.1 mmol)
in dichloromethane (360 mL). The resulting light pink solution was
cooled to -78 °C. A stream of O3 was passed through the solution
at -78 °C until the solution turned colorless. Triphenylphospine
(7.52 g, 28.7 mmol) was added. The solution was allowed to warm
slowly to rt and stirred at rt for 1 h before being concentrated. The
residue was filtered through a short pad of silica gel, washing with
hexanes/dichloromethane/ethyl acetate (15/15/1). The filtrate con-
taining the product was concentrated. The residue was purified by
silica gel chromatography (elution with hexanes/ethyl acetate 20/
1) to give ent-30b (10.98 g, 93%) as a slightly pink oil. The
spectrometric data are in accordance with those of 30b.
calcd 517.2570, obsd 517.2557; [R]22 +22.2 (c 0.76, CHCl3).
D
(+)-(1S,2R,3S)-2-(tert-Butyldimethylsilyloxy)-3-((tert-butyldiphe-
nylsilyloxy)methyl)-3-vinylcyclobutanol (59). Sodium borohydride
(8.6 mg, 0.227 mmol) was added to a solution of 58 (56.2 mg,
0.114 mmol) in 5 mL of MeOH at 0 °C. After 15 min, the reaction
mixture was quenched with saturated NH4Cl solution and diluted
with ether and H2O. The organic layer was washed with brine, dried,
filtered, and concentrated. The residue was purified by flash
chromatography (silica gel, hexanes/ethyl acetate 20/1) to afford
59 as a colorless oil (51.0 mg, 90%): IR (neat, cm-1) 3538, 1471,
1428; 1H NMR (500 MHz, CDCl3) δ 7.69-7.66 (m, 4 H),
7.48-7.39 (m, 6 H), 6.00 (dd, J ) 10.9, 17.7 Hz, 1 H), 5.18 (dd,
J ) 1.4, 10.9 Hz, 1 H), 5.06 (dd, J ) 1.4, 17.7 Hz, 1 H), 4.50 (dd,
J ) 2.3, 5.7 Hz, 1 H), 4.25-4.20 (m, 1 H), 3.60 (d, J ) 10.3 Hz,
1 H), 3.49 (d, J ) 10.3 Hz, 1 H), 2.77 (d, J ) 8.5 Hz, 1 H), 2.28
(ddd, J ) 2.6, 6.7, 10.0 Hz, 1 H), 1.99 (dd, J ) 5.0, 12.5 Hz, 1 H),
1.11 (s, 9 H), 0.94 (s, 9 H), 0.095 (s, 3 H), 0.088 (s, 3 H); 13C
NMR (125 MHz, CDCl3) δ 138.5, 135.70, 135.65, 133.3, 129.7,
127.73, 127.71, 115.2, 72.6, 67.8, 65.4, 49.1, 35.5, 26.9, 25.9, 19.4,
18.3, -4.6, -4.9; HRMS m/z (M + Na)+ calcd 519.2727, obsd
Acknowledgment. This study was supported by The Ohio
State University.
Note Added after ASAP Publication. Due to a production
error, the two paragraphs preceding the section ″Total Synthesis
of (–)-Fomannosin″ were missing in the version published ASAP
May 20, 2008; the corrected version was published ASAP May
30, 2008.
Supporting Information Available: Full experimental de-
tails, high-field 1H and 13C NMR spectra, and full characteriza-
tion for all new compounds described herein. This material is
available free of charge via the Internet at http://pubs.acs.org.
519.2708; [R]22 + 14.8 (c 0.54, CHCl3).
D
JO8004233
(+)-(1S,2R,3S)-2-(tert-Butyldimethylsilyloxy)-1-((tert-butyldi-
phenylsilyloxy)methyl)-3-(4-methoxybenzyloxy)cyclobutane-
carbaldehyde (ent-30b). A solution of TfOH (0.13 µL, 1.45 µmol)
in 50 µL of dry ether was added very slowly into a solution of 59
(49) (a) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483. (b) Bennani, Y. L.; Sharpless, K. B. Tetrahedron Lett. 1993,
34, 2083.
4558 J. Org. Chem. Vol. 73, No. 12, 2008